This study evaluated the effects of the pre-exercise (30 minutes) ingestion of galactose (Gal) or glucose (Glu) on endurance capacity as well as glycaemic and insulinaemic responses. Ten trained male cyclists completed three randomised high-intensity cycling endurance tests. Thirty minutes prior to each trial cyclists ingested 1 litre of either 40g of glucose, 40g of galactose, or a placebo in a double blind manner. The protocol comprised: 20 minutes of progressive incremental exercise (70% to 85% maximal power output (W max )); ten 90 second bouts at 90% W max , separated by 180 seconds at 55% W max ; 90% W max until exhaustion. Blood samples were drawn throughout the protocol. Times to exhaustion were longer with Gal (68.7±10.2 minutes, P=0.005) compared to Glu (58.5±24.9 minutes), with neither being different to placebo (63.9±16.2 minutes). Twenty-eight minutes following Glu consumption, plasma glucose and serum insulin concentrations were higher than with Gal and placebo (P<0.001). Following the initial 20 minutes of exercise, plasma glucose concentrations increased to a relative hyperglycaemia during the Gal and placebo, compared to Glu condition. Higher plasma glucose concentrations during exercise, and the attenuated serum insulin response at rest, may explain the significantly longer times to exhaustion produced by Gal compared to Glu. However, neither carbohydrate treatment produced significantly longer times to exhaustion than placebo, suggesting that the pre-exercise ingestion of galactose and glucose alone is not sufficient to support this type of endurance performance.